Urological and Colorectal Complications Following Surgery for Rectovaginal Endometriosis
Total Page:16
File Type:pdf, Size:1020Kb
DOI: 10.1111/j.1471-0528.2007.01477.x Gynaecological surgery www.blackwellpublishing.com/bjog Urological and colorectal complications following surgery for rectovaginal endometriosis A Slack,a T Child,a I Lindsey,b S Kennedy,a C Cunningham,b N Mortensen,b P Koninckx,a E McVeigha a Nuffield Department of Obstetrics and Gynaecology and b Department of Colorectal Surgery, John Radcliffe Hospital, Oxford, UK Correspondence: Dr E McVeigh, Nuffield Department of Obstetrics and Gynaecology, Level 3, Women’s Centre, John Radcliffe Hospital, Oxford OX3 9DU, UK. Email [email protected] Accepted 24 June 2007. Objectives To report the short- and medium-term complications of Results A total of 128 women underwent surgery. Of these, 32 laparoscopic laser excisional surgery for rectovaginal endometriosis. required intraoperative closure of a rectal wall defect, including 3 segmental rectosigmoid resections. There were three rectovaginal Design Retrospective cohort study. fistulae and one ureterovaginal fistula. Ureteric damage Setting University teaching hospital, UK. occurred in two women, and five women suffered postoperative urinary retention. The risk of intraoperative bowel intervention Population A total of 128 women with histologically confirmed was increased in women who complained of cyclical rectal rectovaginal endometriosis who underwent laparoscopic laser bleeding. surgery between May 1999 and September 2006. Conclusion Laparoscopic laser excision of rectovaginal Methods Women were identified from operative database, and endometriosis is a safe procedure with similar, if not lower, a case note review was performed. Data for surgical outcome and complication rates to other published surgical series. surgical complications were collected. Keywords Complications, endometriosis, rectovaginal, Main outcome measures Rates of urinary tract and colorectal surgery. complications. Please cite this paper as: Slack A, Child T, Lindsey I, Kennedy S, Cunningham C, Mortensen N, Koninckx P, McVeigh E. Urological and colorectal complications following surgery for rectovaginal endometriosis. BJOG 2007;114:1278–1282. Introduction Medical therapy, which has been shown to be effective in the symptomatic treatment of superficial disease, has little ef- Endometriosis is characterised by the presence of glandular fect on deep rectovaginal endometriosis.7–9 Surgery for recto- and stromal tissues in areas outside the uterus. It has been vaginal endometriosis can be complex and challenging and considered for decades as the result of the implantation often involves a multidisciplinary team (MDT). The choice of of retrograde menstruated endometrial cells1 (Sampson’s surgery will depend to some extent on the reproductive plans theory) or as metaplasia2,3 induced by menstrual debris or of the women as well as the surgical skill available. Many as lymphatic spread.4,5 It occurs most frequently in the pelvic authors, us included, advocate the complete dissection and organs and peritoneum and is prevalent in 2.5–3.3% of surgical removal of the affected tissues.10–13 There is, however, women in the reproductive age. no consensus as to the best way of achieving this and many Rectovaginal endometriosis accounts for 5–10% of women differing approaches have been described,10,11,14,15 none of with endometriosis and is increasingly becoming recognised which has been accepted as ‘best’ practice. The main area of as a separate clinical syndrome from the superficial form of uncertainty is how to treat the affected bowel. One approach the disease.6 It is characterised by the presence of palpable is conservative, with partial or complete excision of the endo- endometriotic nodules deep in the connective tissue of the metriotic lesion without rectal resection. Other authors advo- pelvis, which show profound fibrosis and fibromuscular cate a more radical approach involving segmental rectal hyperplasia. Women with this form of the disease usually resection. To help develop a consensus regarding the ideal have severe symptoms, and there are significant risks of uri- surgical approach, it is essential to closely monitor the efficacy nary tract and bowel involvement. and complications of different surgical approaches. 1278 ª 2007 The Authors Journal compilation ª RCOG 2007 BJOG An International Journal of Obstetrics and Gynaecology Rectovaginal endometriosis The purpose of this study is to report the outcomes of mortality, hospital stay duration and follow-up period. a series of women with rectovaginal endometriosis undergo- Descriptive statistics were calculated and tabulated. ing conservative excisional treatment by a MDT of gynaeco- logists and colorectal surgeons. Results Methods Patient demographics From May 1999 to September 2006, 495 women with severe A retrospective analysis of all women undergoing laparo- revised American Fertility Society stage IV endometriosis scopic surgery for complex rectovaginal endometriosis were identified. A total of 128 women had histological con- between May 1999 and September 2006 was undertaken. firmation of rectovaginal endometriosis, and this formed the Eligible women were identified from operative databases study cohort. The mean age of women was 32 years, with and by examining case notes. The study cohort was restricted a range of 22–47 years. to women with histological confirmation of rectovaginal endo- metriosis characterised by endometrial-like glands and stroma Symptoms surrounded by much fibrosis and smooth-muscle hyperplasia Presenting symptoms were available for 104 women and are in the rectovaginal septum. described in Table 1. All women were extensively counselled about the risks of surgery, which was undertaken without hormonal pre- Surgery treatment. Preoperative patient assessment depended on The mean operating time was 106 minutes (range 35–240 symptoms. All women underwent vaginal and rectal exami- minutes). This did not alter over the period of the study, with nation performed preferentially at the time of menstruation. the mean operating time of the first half of the study 105 If there was a significant history of dyschezia or menstrual minutes (35–240 minutes) and the second half 109 minutes rectal bleeding, then magnetic resonance imaging and/or bar- (60–240 minutes). ium enema study were performed. In cases of ureteric All cases were undertaken laparoscopically, and no women involvement with hydronephrosis, ureteric stents were required intraoperative conversion to laparotomy. Three inserted prior to surgery. A significant proportion of women women required a mini extension of one lateral port site to had been referred for tertiary care following laparoscopic remove the specimen and to facilitate a stapled anastomosis assessment by another gynaecologist. following bowel resection. A rectal wall defect required suture All women had 24 hours bowel preparation. Surgery was closure in 32 women (25%), in 17 women (14%) to repair carried out using a carbon dioxide laser. The surgical tech- a full-thickness defect and in 15 women (11%) a partial defect nique began with lysis of adhesions, drainage and stripping of of the muscularis. One woman required resection of a segment ovarian endometriomas and identification of both ureters in of rectosigmoid colon due to extensive full-thickness disease the pelvic sidewall. Both pararectal spaces were identified and and the presence of diverticulitis, and two women required opened below the extent of the disease to isolate the endo- segmental rectosigmoid resection due to the presence of a cir- metriotic nodules. The nodule was then dissected away from cumferential lesion. the bowel until healthy tissue was reached, and finally, it was The surgery was carried out by gynaecologists with assis- dissected free from the rectovaginal septum, including an tance from colorectal surgeons when required (Table 2). excision of the vaginal vault if necessary, which was then closed laparoscopically. All specimens were sent for histolog- Risk of bowel surgery versus symptoms ical examination. The risk of bowel intervention was increased in women with Rectal wall defects resulting from excision of lesions cyclical rectal bleeding. A full-thickness rectal wall defect extending into (partial) or through (full thickness) the mus- required closing in 9 of 17 women (53%) with cyclical rectal cularis propria of the rectum were closed laparoscopically in bleeding compared with 7 of 87 (8%) women without cyclical two layers with vicryl sutures. At the end of the procedure, rectal bleeding. A partial-thickness defect required closing in meticulous haemostasis was achieved, and the integrity of the bowel was checked by an underwater air leak test, supple- mented in the later part of the series by installation of Table 1. Symptoms Betadine into the rectum to look for suture line leakage. All cases were recorded on either VHS video, CD-ROM or Dyspareunia 77 (74%) DVD. Women received a completed copy of the recording. Dysmenorrhoea 81 (78%) The following data were retrieved from the surgical data- Dyschezia 40 (38%) base: age, preoperative symptoms, operative approach, type of Rectal bleeding 17 (16%) rectal surgery, histopathology, operative morbidity and ª 2007 The Authors Journal compilation ª RCOG 2007 BJOG An International Journal of Obstetrics and Gynaecology 1279 Slack et al. Table 2. Type of surgery Table 4. Complications Gynaecologist Gynaecologist Urinary retention 5 (3.9%) only and Ureteric damage 1 (0.8%) colorectal Rectovaginal fistula 3 (2.3%) surgeon Uretovaginal fistula 1 (0.8%) Complete discoid resection